Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2...

28
1 Formation mechanisms of atmospheric nitrate and sulfate during the 1 winter haze pollution periods in Beijing: gas-phase, heterogeneous 2 and aqueous-phase chemistry 3 Pengfei Liu 1, 2, 3, 5 , Can Ye 1, 3 , Chaoyang Xue 1, 3 , Chenglong Zhang 1, 2, 3 , Yujing Mu 1, 2, 3, 4 , Xu 4 Sun 1, 6 5 1 Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, 100085, China. 6 2 Center for Excellence in Urban Atmospheric Environment, Institute of Urban Environment, Chinese Academy of 7 Sciences, Xiamen, 361021, China. 8 3 University of Chinese Academy of Sciences, Beijing, 100049, China. 9 4 National Engineering Laboratory for VOCs Pollution Control Material & Technology, University of Chinese 10 Academy of Sciences, Beijing, 100049, China. 11 5 Key Laboratory of Atmospheric Chemistry, China Meteorological Administration, Beijing, 100081, China. 12 6 Beijing Urban Ecosystem Research Station, Beijing, 100085, China. 13 Correspondence: Yujing Mu ([email protected]) 14 Abstract 15 A vast area in China is currently going through severe haze episodes with drastically elevated 16 concentrations of PM2.5 in winter. Nitrate and sulfate are main constituents of PM2.5 but their 17 formations via NO2 and SO2 oxidation are still not comprehensively understood, especially under 18 different pollution or atmospheric relative humidity (RH) conditions. To elucidate formation 19 pathways of nitrate and sulfate in different polluted cases, hourly samples of PM2.5 were collected 20 continuously in Beijing during the wintertime of 2016. Three serious pollution cases were 21 identified reasonably during the sampling period and the secondary formations of nitrate and 22 sulfate were found to make a dominant contribution to atmospheric PM2.5 under the relatively high 23 RH condition. The significant correlation between NOR (NOR = NO3 - / (NO3 - +NO2)) and [NO2] 2 × 24 [O3] during the nighttime under the RH≥60% condition indicated that the heterogeneous 25 hydrolysis of N2O5 involving aerosol liquid water was responsible for the nocturnal formation of 26 nitrate at the extremely high RH levels. The more coincident trend of NOR and [HONO] × [DR] 27

Transcript of Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2...

Page 1: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

1

Formation mechanisms of atmospheric nitrate and sulfate during the 1

winter haze pollution periods in Beijing: gas-phase, heterogeneous 2

and aqueous-phase chemistry 3

Pengfei Liu1, 2, 3, 5, Can Ye1, 3, Chaoyang Xue1, 3, Chenglong Zhang1, 2, 3, Yujing Mu1, 2, 3, 4, Xu 4

Sun1, 6 5

1 Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, 100085, China. 6

2 Center for Excellence in Urban Atmospheric Environment, Institute of Urban Environment, Chinese Academy of 7

Sciences, Xiamen, 361021, China. 8

3 University of Chinese Academy of Sciences, Beijing, 100049, China. 9

4 National Engineering Laboratory for VOCs Pollution Control Material & Technology, University of Chinese 10

Academy of Sciences, Beijing, 100049, China. 11

5 Key Laboratory of Atmospheric Chemistry, China Meteorological Administration, Beijing, 100081, China. 12

6 Beijing Urban Ecosystem Research Station, Beijing, 100085, China. 13

Correspondence: Yujing Mu ([email protected]) 14

Abstract 15

A vast area in China is currently going through severe haze episodes with drastically elevated 16

concentrations of PM2.5 in winter. Nitrate and sulfate are main constituents of PM2.5 but their 17

formations via NO2 and SO2 oxidation are still not comprehensively understood, especially under 18

different pollution or atmospheric relative humidity (RH) conditions. To elucidate formation 19

pathways of nitrate and sulfate in different polluted cases, hourly samples of PM2.5 were collected 20

continuously in Beijing during the wintertime of 2016. Three serious pollution cases were 21

identified reasonably during the sampling period and the secondary formations of nitrate and 22

sulfate were found to make a dominant contribution to atmospheric PM2.5 under the relatively high 23

RH condition. The significant correlation between NOR (NOR = NO3-/ (NO3

-+NO2)) and [NO2]2 × 24

[O3] during the nighttime under the RH≥60% condition indicated that the heterogeneous 25

hydrolysis of N2O5 involving aerosol liquid water was responsible for the nocturnal formation of 26

nitrate at the extremely high RH levels. The more coincident trend of NOR and [HONO] × [DR] 27

Page 2: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

2

(direct radiation) × [NO2] than [Dust] × [NO2] during the daytime under the 30%<RH<60% 28

condition provided convincing evidence that the gas-phase reaction of NO2 with OH played a 29

pivotal role in the diurnal formation of nitrate at moderate RH levels. The extremely high mean 30

values of SOR (SOR = SO42- / (SO4

2-+SO2)) during the whole day under the RH≥60% condition 31

could be ascribed to the evident contribution of SO2 aqueous-phase oxidation to the formation of 32

sulfate during the severe pollution episodes. Based on the parameters measured in this study and 33

the known sulfate production rate calculation method, the oxidation pathway of H2O2 rather than 34

NO2 was found to contribute greatly to the aqueous-phase formation of sulfate. 35

1. Introduction 36

In recent years, severe haze has occurred frequently in Beijing as well as the North China 37

Plain (NCP) during the wintertime, which has aroused great attention from the public due to its 38

adverse impact on atmospheric visibility, air quality and human health (Chan and Yao, 39

2008;Zhang et al., 2012;Zhang et al., 2015). 40

To mitigate the severe haze pollution situations, a series of regulatory measures for primary 41

pollution sources have been implemented by the Chinese government. For example, coal 42

combustion for heating in winter has gradually been replaced with electricity and natural gas in 43

the NCP, coal-fired power plants have been strictly required to install flue-gas denitration and 44

desulfurization systems (Chen et al., 2014), the stricter control measures such as terminating 45

production in industries and construction as well as the odd and even number rule for vehicles 46

have been performed in megacities during the period of the red alert for haze and so on. These 47

actions have made tremendous effects to decline pollution levels of primary pollutants including 48

PM2.5 (fine particulate matter with an aerodynamic diameter less than 2.5 m) in recent years (Li 49

Page 3: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

3

et al., 2019). However, the serious pollution events still occurred in many areas of 50

Beijing-Tianjin-Hebei (BTH) region in December 2016 and January 2017 (Li et al., 2019). It has 51

been acknowledged that the severe haze pollution is mainly ascribed to stagnant meteorological 52

conditions with high atmospheric relative humidity (RH) and low mixed boundary layer height, 53

strong emissions of primary gaseous pollutants and rapid formation of secondary inorganic 54

aerosols (SIAs, the sum of sulfate, nitrate and ammonium), especially sulfate and nitrate (Cheng et 55

al., 2016;Guo et al., 2014;Huang et al., 2014). Some studies suggested that the contribution of 56

SIAs to PM2.5 was higher than 50% during the most serious haze days (Quan et al., 2014;Xu et al., 57

2017;Zheng et al., 2015a). 58

Generally, atmospheric sulfate and nitrate are formed through the oxidations of the precursor 59

gases (SO2 and NO2) by oxidants (e.g. OH radical, O3) via gas-phase, heterogeneous and 60

aqueous-phase reactions (Ravishankara, 1997;Wang et al., 2013;Yang et al., 2015). It should be 61

noted that the recent study proposed the remarkable emissions of primary sulfate from residential 62

coal combustion with the sulfur contents of coal in range of 0.81-1.88% in Xi’an (Dai et al., 2019), 63

but the primary emissions of sulfate could be neglected due to the extremely low sulfur content of 64

coal (0.26-0.34%) used prevailingly in the NCP (Du et al., 2016;Li et al., 2016). Atmospheric RH 65

is a key factor that facilitates the SIAs formation and aggravates the haze pollution (Wu et al., 66

2019), and hence the secondary formations of sulfate and nitrate are simply considered to be 67

mainly via gas-phase reaction at relatively low atmospheric RH levels (RH<30%) and 68

heterogeneous reactions and aqueous-phase reactions at relatively high atmospheric RH levels 69

(RH>60%) (Li et al., 2017). However, their formation mechanisms at different atmospheric RH 70

levels still remain controversial and unclear (Cheng et al., 2016;Ge et al., 2017;Guo et al., 2017;Li 71

Page 4: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

4

et al., 2018;Liu et al., 2017a;Wang et al., 2016;Yang et al., 2017). For example, the recent studies 72

proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 73

extremely high atmospheric RH conditions played a dominant role in sulfate formation under 74

almost neutral aerosol solutions (a pH range of 5.4-7.0) during the serious pollution periods 75

(Cheng et al., 2016;Wang et al., 2018a;Wang et al., 2016). However, Liu et al. (2017a) and Guo et 76

al. (2017) found that the aerosol pH estimated by ISORROPIA-II model was moderately acidic (a 77

pH range of 3.0-4.9) and thus the pathway of SO2 aqueous-phase oxidation by dissolved NO2 was 78

unimportant during severe haze events in China. Additionally, although the pathway of N2O5 79

heterogeneous hydrolysis has been recognized as being responsible for the nocturnal formation of 80

NO3- under relatively high atmospheric RH conditions (Tham et al., 2018;Wang et al., 81

2018b;Wang et al., 2018c), the effects of NO2 gas-phase chemistry and NO2 heterogeneous 82

chemistry on the diurnal formation of NO3- under moderate atmospheric RH conditions 83

(30%<RH<60%) have not yet been understood. Therefore, measurements of the species in PM2.5 84

in different polluted cases during the wintertime are urgently needed to elucidate formation 85

pathways of sulfate and nitrate. 86

In this study, hourly filter samples of PM2.5 were collected continuously in Beijing during the 87

wintertime of 2016, and the pollution characteristics and formation mechanisms of sulfate and 88

nitrate in the PM2.5 samples were investigated comprehensively under different atmospheric RH 89

conditions in relation to gas-phase, heterogeneous and aqueous-phase chemistry. 90

2. Materials and Methods 91

2.1. Sampling and analysis 92

The sampling site was chosen on the rooftop (around 25 m above the ground) of a six-story 93

Page 5: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

5

building in Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences 94

(RCEES, CAS), which was located in the northwest of Beijing and had been described in detail by 95

our previous studies (Liu et al., 2016;Liu et al., 2017b). The location of the sampling site 96

(400029.85 N, 1162029.71 E) is presented in Figure S1. Hourly PM2.5 samples were collected 97

on prebaked quartz fiber filters (90mm, Munktell) from January 7th to 23th of 2016 by 98

median-volume samplers (Laoying-2030) with a flow rate of 100 L min-1. Water-soluble ions 99

(WSI), including Na+, NH4+, Mg2+, Ca2+, K+, Cl-, NO2

-, NO3- and SO4

2-, as well as carbon 100

components including organic carbon (OC) and element carbon (EC) in the filter samples were 101

analyzed by ion chromatography (Wayeal IC6200) and thermal optical carbon analyzer 102

(DRI-2001A), respectively (Liu et al., 2017b). Analysis relevant for quality assurance & quality 103

control (QA/QC) was presented in detail in section M1 of the Supplementary Information (SIs). 104

Atmospheric H2O2 and HONO were monitored by AL2021-H2O2 monitor (AERO laser, Germany) 105

and a set of double-wall glass stripping coil sampler coupled with ion chromatography (SC-IC), 106

respectively (Ye et al., 2018;Xue et al., 2019a;Xue et al., 2019b). More details about the 107

measurements of H2O2 and HONO were ascribed in section M2 of the SIs. Meteorological data, 108

including wind speed, wind direction, ambient temperature and RH, as well as air quality index 109

(AQI) derived by PM2.5, SO2, NOx, CO and O3 were obtained from Beijing urban ecosystem 110

research station in RCEES, CAS (http://www.bjurban.rcees.cas.cn/). 111

2.2. Aerosol liquid water contents and pH prediction by ISORROPIA-II model 112

The ISORROPIA-II model was employed to calculate the equilibrium composition for 113

Na+-K+-Ca2+-Mg2+-NH4+-Cl--NO3

--SO42--H2O aerosol system, which is widely used in regional 114

and global atmospheric models and has been successfully applied in numerous studies for 115

Page 6: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

6

predicting the physical state and composition of atmospheric inorganic aerosols (Fountoukis and 116

Nenes, 2007;Guo et al., 2015;Shi et al., 2017). It can be used in two modes: forward mode and 117

reverse mode. Forward mode calculates the equilibrium partitioning given the total concentrations 118

of gas and aerosol species, whereas reverse mode involves predicting the thermodynamic 119

compositions based only on the concentrations of aerosol components. Forward mode was 120

adopted in this study because reverse mode calculations have been verified to be not suitable to 121

characterize aerosol acidity (Guo et al., 2015;Hennigan et al., 2015;Murphy et al., 2017;Pathak et 122

al., 2004;Weber et al., 2016). The ISORROPIA-II model is available in “metastable” or “solid + 123

liquid” state solutions. Considering the relatively high RH during the sampling period, the 124

metastable state solution was selected in this study due to its better performance than the latter 125

(Bougiatioti et al., 2016;Guo et al., 2015;Liu et al., 2017a;Weber et al., 2016). Additionally, 126

although the gaseous HNO3, H2SO4, HCl and NH3 were not measured in this study, gas-phase 127

input with the exception of NH3 has an insignificant impact on the aerosol liquid water contents 128

(ALWC) and pH calculation due to the lower concentrations of HNO3, H2SO4 and HCl than NH3 129

in the atmosphere (Ding et al., 2019;Guo et al., 2017). Based on the long-term measurement in the 130

winter of Beijing, an empirical equation between NOx and NH3 concentrations was derived from 131

the previous study (Meng et al., 2011), that is, NH3 (ppb) = 0.34 × NOx (ppb) + 0.63, which was 132

employed for estimating the NH3 concentration in this study. The predicted daily average 133

concentrations of NH3 varied from 3.3 g m-3 to 36.9 g m-3, with a mean value of 16.6 g m-3 134

and a median value of 14.6 g m-3, which were in line with those (7.6-38.1 g m-3, 18.2 g m-3 135

and 16.2 g m-3 for the daily average concentrations, the mean value and the median value of NH3, 136

respectively) during the winter of 2013 in Beijing in the previous study (Zhao et al., 2016). 137

Page 7: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

7

Then, the aerosol pH could be calculated by the following equation: 138

𝑝𝐻 = −𝑙𝑜𝑔101000×𝐻+

𝑊 139

where H+ (g m-3) and W (g m-3) are the equilibrium particle hydrogen ion concentration 140

and aerosol water contents, respectively, both of which could be output from ISORROPIA-II. 141

2.3. Production of sulfate in aqueous-phase reactions 142

The previous studies showed that there were six pathways of the aqueous-phase oxidation of 143

SO2 to the production of sulfate, i.e. H2O2 oxidation, O3 oxidation, NO2 oxidation, transition metal 144

ions (TMI) + O2 oxidation, methyl hydrogen peroxide (MHP) oxidation and peroxyacetic acid 145

(PAA) oxidation (Cheng et al., 2016;Zheng et al., 2015a). Because some TMIs, such as Ti(III), 146

V(III), Cr(III), Co(II), Ni(II), Cu(II) and Zn(II), displayed much less catalytic activities (Cheng et 147

al., 2016), only Fe(III) and Mn(II) were considered in this study. In addition, due to the extremely 148

low concentrations of MHP and PAA in the atmosphere, their contributions to the production of 149

sulfate could be ignored (Zheng et al., 2015a). To investigate the formation mechanism of sulfate 150

during the serious pollution episodes, the contributions of O3, H2O2, NO2 and Fe(III) + Mn(II) to 151

the production of sulfate in aqueous-phase reactions were calculated by the formulas as follows 152

(Cheng et al., 2016;Ibusuki and Takeuchi, 1987;Seinfeld and Pandis, 2006): 153

−(𝑑[𝑆(𝐼𝑉)]

𝑑𝑡)𝑂3 = (𝑘0[𝑆𝑂2𝐻2𝑂] + 𝑘1[𝐻𝑆𝑂3

−] + 𝑘2[𝑆𝑂32−])[𝑂3(𝑎𝑞)] (R1) 154

−(𝑑[𝑆(𝐼𝑉)]

𝑑𝑡)𝐻2𝑂2 =

𝑘3[𝐻+][𝐻𝑆𝑂3

−][𝐻2𝑂2(𝑎𝑞)]

1+𝐾[𝐻+] (R2) 155

−(𝑑[𝑆(𝐼𝑉)]

𝑑𝑡)𝐹𝑒(𝐼𝐼𝐼)+𝑀𝑛(𝐼𝐼) = 𝑘4[𝐻

+]𝑎[𝑀𝑛(𝐼𝐼)][𝐹𝑒(𝐼𝐼𝐼)][𝑆(𝐼𝑉)] (R3) 156

−(𝑑[𝑆(𝐼𝑉)]

𝑑𝑡)𝑁𝑂2 = 𝑘5[𝑁𝑂2(𝑎𝑞)][𝑆(𝐼𝑉)] (R4) 157

where k0 = 2.4×104 M-1 s-1, k1 = 3.7×105 M-1 s-1, k2 = 1.5×109 M-1 s-1, k3 = 7.45×107 M-1 s-1, K = 158

13 M-1, k4 = 3.72×107 M-1 s-1, a = -0.74 (pH≤4.2) or k4 = 2.51×1013 M-1 s-1, a = 0.67 (pH>4.2), and 159

Page 8: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

8

k5 = (1.24-1.67)×107 M-1 s-1 (5.3≤pH≤8.7, the linear interpolated values were used for pH between 160

5.3 and 8.7) at 298 K (Clifton et al., 1988); [O3(aq)], [H2O2(aq)] and [NO2(aq)] could be calculated by 161

the Henry’s constants which are 1.1×10-2 M atm-1, 1.0×105 M atm-1 and 1.0×10-2 M atm-1 at 298 K 162

for O3, H2O2 and NO2 respectively (Seinfeld and Pandis, 2006). As for [Fe(III)] and [Mn(II)], their 163

concentrations entirely depended on the values of pH due to the precipitation equilibriums of 164

Fe(OH)3 and Mn(OH)2 (Graedel and Weschler, 1981). Considering the aqueous-phase ionization 165

equilibrium of SO2, the Henry’s constants of HSO3-, SO3

2- and S(IV) could be expressed by the 166

equations as follows (Seinfeld and Pandis, 2006): 167

𝐻𝐻𝑆𝑂3−∗ = 𝐻𝑆𝑂2

𝐾𝑆1

[𝐻+] (R5) 168

𝐻𝑆𝑂32−∗ = 𝐻𝑆𝑂2

𝐾𝑆1𝐾𝑆2

[𝐻+]2 (R6) 169

𝐻𝑆(𝐼𝑉)∗ = 𝐻𝑆𝑂2(1 +

𝐾𝑆1

[𝐻+]+

𝐾𝑆1𝐾𝑆2

[𝐻+]2) (R7) 170

where HSO2 = 1.23 M atm-1, KS1 = 1.3×10-2 M and KS2 = 6.6×10-8 M at 298 K. In addition, all 171

of rate constants (k), Henry’s constants (H) and ionization constants (K) are evidently influenced 172

on the ambient temperature and are calibrated by the formulas as follows (Seinfeld and Pandis, 173

2006): 174

𝑘(𝑇) = 𝑘(𝑇0)𝑒[−

𝐸

𝑅(1

𝑇−

1

𝑇0)]

(R8) 175

𝐻(𝑇) = 𝐻(𝑇0)𝑒[−

∆𝐻

𝑅(1

𝑇−

1

𝑇0)]

(R9) 176

𝐾(𝑇) = 𝐾(𝑇0)𝑒[−

𝐸

𝑅(1

𝑇−

1

𝑇0)]

(R10) 177

where T is the ambient temperature, T0=298 K, both E/R and ∆H/R varied in the different 178

equations and their values could be found in Cheng et al., (2016). 179

Furthermore, mass transport was also considered for multiphase reactions in different 180

medium and across the interface in section M3 of the SIs. 181

Page 9: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

9

3. Results and Discussion 182

3.1. Variation characteristics of the species in PM2.5 and typical gaseous pollutants 183

The concentrations of the species in PM2.5 and typical gaseous pollutants including NO2, SO2, 184

O3, HONO and H2O2 as well as atmospheric RH are shown in Figure 1. The meteorological 185

parameters (wind speed, wind direction, ambient temperature and direct radiation (DR)) as well as 186

the concentrations of PM2.5 are displayed in Figure S2. During the sampling period, the 187

concentrations of the species in PM2.5 and typical gaseous pollutants varied similarly on a 188

timescale of hours with a distinct periodic cycle of 3-4 days, suggesting that meteorological 189

conditions played a vital role in accumulation and dispersion of atmospheric pollutants (Xu et al., 190

2011;Zheng et al., 2015b). For example, the relatively high levels of PM2.5 (>100 g m-3) usually 191

occurred under the relatively stable meteorological conditions with the low south wind speed (<2 192

m s-1) and the high RH (>60%) which favored the accumulation of pollutants. Besides 193

meteorological conditions, the extremely high concentrations of the species in PM2.5 might be 194

mainly ascribed to strong emissions of primary pollutants and rapid formation of secondary 195

aerosols during the wintertime in Beijing. 196

The average concentrations of the species in PM2.5 and typical gaseous pollutants during 197

clean or slightly polluted (C&SP) episodes (PM2.5<75 g m-3), during polluted or heavy polluted 198

(P&HP) episodes (PM2.5≥75 g m-3) and during the whole sampling period are illustrated in Table 199

1. It is evident that the average concentrations of NO3-, SO4

2-, NH4+, OC and EC during P&HP 200

episodes were about a factor of 5.0, 4.1, 6.1, 3.6 and 3.2 greater than those during C&SP episodes, 201

respectively, indicating that the formations of SIAs were more efficient compared to other species 202

in PM2.5 during the serious pollution episodes. Given that the average concentrations of gaseous 203

Page 10: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

10

precursors (NO2 and SO2) during P&HP episodes were approximately a factor of 2.0-2.2 greater 204

than those during C&SP episodes, the obviously higher elevation of NO3- and SO4

2- implied that 205

the oxidations of NO2 and SO2 by the major atmospheric oxidizing agents (OH radicals, O3 and 206

H2O2 etc.) might be greatly accelerated due to the relatively high concentrations of oxidants and 207

atmospheric RH during the serious pollution episodes (Figure 1). The average concentration of 208

H2O2 was found to be a factor of 1.7 greater during P&HP episodes than during C&SP episodes, 209

indicating that atmospheric H2O2 might contribute to the formation of SIAs especially sulfate 210

during the serious pollution episodes with high atmospheric RH, which will be discussed in Sect. 211

3.3.2. However, the obvious decrease in O3 average concentration was observed during P&HP 212

episodes compared to C&SP episodes, which was mainly attributed to the relatively weak solar 213

radiation and the titration of NO during the serious pollution episodes (Ye et al., 2018). In addition, 214

the evidently higher average concentration of HONO during P&HP episodes than during C&SP 215

episodes might be also due to the relatively weak solar radiation as well as the heterogeneous 216

reaction of NO2 on particle surfaces during the serious pollution episodes (Tong et al., 2016;Wang 217

et al., 2017). 218

3.2. Three serious pollution cases during the sampling period 219

Based on the transition from the clean to polluted periods, three haze cases were identified 220

during the sampling period (Figure 1 and Figure S2): from 13:00 on January 8th to 1:00 on January 221

11th (Case 1), from 14:00 on January 14th to 7:00 on January 17th (Case 2), and from 8:00 on 222

January 19th to 2:00 on January 22nd (Case 3). The serious pollution duration in the three cases 223

could last 1-3 days due to the differences of their formation mechanisms. 224

In Case 1, the variation trends of the concentrations of the species in PM2.5, NO2, SO2, 225

Page 11: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

11

HONO and H2O2 were almost identical and exhibited three pollution peaks at night (Figure 1), 226

which might be ascribed to the possibility that the decrease of nocturnal mixed boundary layer 227

accelerated the pollutant accumulation (Bei et al., 2017;Zhong et al., 2019). Considering the 228

relatively low RH (15-40%) and wind speeds (<2 m s-1) in Case 1 (Figure S2), primary emissions 229

around the sampling site were suspected to be a dominant source for the increase in the PM2.5 230

concentrations. Further evidence is that the correlation between the concentrations of PM2.5 and 231

CO is better in Case 1 (R2=0.55) than in Case 2 and Case 3 (R2=0.20~0.52) (Figure S3). Identical 232

to Case 1, three obvious pollution peaks were also observed in Case 2 (Figure 1). The variation 233

trends of the concentrations of the species in PM2.5 and typical gaseous pollutants at the first peak 234

in Case 2 were found to be similar with those in Case 1, which were mainly attributed to their 235

similar formation mechanism. However, the evident decreases in NOx and SO2 were observed 236

when the concentrations of the species in PM2.5 were increasing and the atmospheric oxidation 237

pollutant (e.g. H2O2) concentration peaks were prior to others at the last two peaks in Case 2, 238

suggesting that secondary formation from gaseous precursors might be dominant for PM2.5 239

pollution. The relatively high RH (50-80%) and the low south wind speeds (<2 m s-1) in Case 2 240

(Figure S2) provided further evidence for the above speculation. In contrast to Case 1 and Case 2, 241

the relatively high south wind speeds (>3 m s-1) (Figure S2) with the concentrations of the species 242

in PM2.5 and typical gaseous pollutants increasing slowly (Figure 1) at the beginning of Case 3 243

indicated that regional transportation might be responsible for the atmospheric species. 244

Subsequently, the concentrations of the species in PM2.5 remained relatively high when the 245

atmospheric RH lasted more than 60%, implying that secondary formation from gaseous 246

precursors dominated PM2.5 pollution during the late period of Case 3. 247

Page 12: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

12

The average mass proportions of the species in PM2.5 in the three cases are illustrated in 248

Figure S4, the proportions of the primary species such as EC (10-13%), Cl- (6-7%) and Na+ (4%) 249

in the three cases were almost identical, indicating that primary particle emissions were relatively 250

stable during the sampling period. However, the proportions of SIA in Case 2 (42%) and Case 3 251

(38%) were conspicuously greater than that in Case 1 (28%), further confirming that secondary 252

formation of inorganic ions (e.g. nitrate, sulfate) made a significant contribution to atmospheric 253

PM2.5 in Case 2 and Case 3. 254

3.3. Formation mechanism of nitrate and sulfate during serious pollution episodes 255

As for nitrate and sulfate in the three cases, the highest mass proportion (18%) of nitrate was 256

observed in Case 2, whereas the highest mass proportion (15%) of sulfate was found in Case 3 257

(Figure S4). Although the concentrations of SO2 were obviously lower than the concentrations of 258

NO2 in both Case 2 and Case 3 (Figure 1 and Table 1), the extremely high proportion of sulfate in 259

Case 3 might be ascribed to the long-lasting plateau of RH (Figure S2) because the aqueous-phase 260

reaction could accelerate the conversion of SO2 to SO42-. To further investigate the pollution 261

characteristics of nitrate and sulfate during the serious pollution episodes, the relations between 262

NOR (NOR = NO3- / (NO3

-+NO2)) as well as SOR (SOR = SO42- / (SO4

2-+SO2)) and RH are 263

shown in Figure 2. There were obvious differences in the variations of NOR and SOR under 264

different atmospheric RH conditions. The variation trends of NOR and SOR almost stayed the 265

same when atmospheric RH was below 30%, and then simultaneously increased with atmospheric 266

RH in the range of 30-60%. The enhanced gas-phase reaction and the heterogeneous reaction 267

involving aerosol liquid water might make a remarkable contribution to the elevation of NOR and 268

SOR, respectively, which were further discussed in the following section. Subsequently, the 269

Page 13: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

13

variation trend of NOR slowly decreased whereas the variation trend of SOR significantly 270

increased when atmospheric RH was above 60%, which was very similar with the previous studies 271

(Sun et al., 2013;Zheng et al., 2015b). Considering that the heterogeneous reactions of NO2 on 272

particle surface were dependent on atmospheric RH due to the competition of water for surface 273

reactive sites of particles (Ponczek et al., 2019), the slow reduction of NOR might be due to the 274

suppressed heterogeneous reaction of NO2 to nitrate formation under high RH condition (Tang et 275

al., 2017), while the elevation of SOR revealed the dominant contribution of the aqueous-phase 276

reaction to sulfate formation. 277

3.3.1. Formation mechanism of nitrate 278

Atmospheric nitrate is considered to be mainly from NO2 oxidation by OH radical in the gas 279

phase, heterogeneous uptake of NO2 on the surface of particles and heterogeneous hydrolysis of 280

N2O5 on wet aerosols or chloride-containing aerosols (He et al., 2014;He et al., 2018;Nie et al., 281

2014;Ravishankara, 1997;Wang et al., 2018b). Since atmospheric N2O5 is usually produced by the 282

reaction of NO3 radical with NO2 as well as both NO3 radical and N2O5 are easily photolytic 283

during the daytime, the heterogeneous hydrolysis of N2O5 is a nighttime pathway for the 284

formation of atmospheric nitrate (He et al., 2018;Wang et al., 2018b). As shown in Figure 3a, the 285

mean values of NOR during the nighttime remarkably elevated with atmospheric RH increasing, 286

the disproportionation of NO2 and the heterogeneous hydrolysis of N2O5 involving aerosol liquid 287

water were suspected to dominate the nocturnal formation of nitrate under high RH conditions 288

during the sampling period (Ma et al., 2017;Wang et al., 2018b;Li et al., 2018). However, the 289

productions of HONO and nitrate should be equal through the disproportionation of NO2 (Ma et 290

al., 2017), which could not explain the wide gaps between the average concentrations of HONO 291

Page 14: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

14

(about 6.5 g m-3) and nitrate (about 20.1 g m-3) observed at the nighttime under high RH 292

conditions during the sampling period. Thus, the disproportionation of NO2 made insignificant 293

contribution to the nocturnal formation of nitrate under high RH conditions. Considering that 294

atmospheric NO3 radical is mainly generated via the oxidation of NO2 by O3, the relatively high 295

O3 and NO2 levels could be in favor of the formation of N2O5 during the nighttime (He et al., 296

2018;Wang et al., 2018b), and hence the correlation between [NO2]2 × [O3] and NOR can 297

represent roughly the contribution of the heterogeneous hydrolysis of N2O5 to atmospheric nitrate 298

at night. As shown in Figure 3b, although the variations of [NO2]2 × [O3] at the nighttime 299

(18:00-7:00) were all positively correlated with NOR under the three different RH conditions, 300

their correlation under the RH60% condition (R2=0.552) was significantly stronger than those 301

under the RH<60% condition (R20.181). It has been acknowledged that the correlation between 302

two species means the impact of changes in one species on another. The stronger the correlation is, 303

the greater the impact is. Therefore, the significantly stronger correlations between NOR and 304

[NO2]2 × [O3] under the RH60% condition than under the RH<60% condition revealed that the 305

heterogeneous hydrolysis of N2O5 made a remarkable contribution to atmospheric nitrate at the 306

nighttime under high RH condition. Additionally, the obviously lower slope of the correlation 307

between NOR and [NO2]2 × [O3] under the RH60% condition (slope=11691) than under the 308

RH<60% condition (slope17399) (Figure 3b) also suggested that the formation of atmospheric 309

nitrate at the nighttime under high RH condition was more sensitive to the pathway of N2O5. 310

However, the obvious increase in the mean values of NOR during the daytime (especially for 311

10:00-17:00) under the 30%<RH<60% condition (Figure 3a) indicated that additional sources 312

rather than the heterogeneous hydrolysis of N2O5 were responsible for the formation of nitrate. To 313

Page 15: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

15

explore the possible formation mechanisms of nitrate in this case, the daily variations of [Dust] 314

(the sum of Ca2+ and Mg2+) × [NO2] and [HONO] (the main source of OH) × [DR] × [NO2], 315

which can represent roughly the heterogeneous reaction of NO2 on the surface of mineral aerosols 316

and the gas-phase reaction of NO2 with OH, are shown in Figure 3c and Figure 3d, respectively. 317

The mean values of [HONO] × [DR] × [NO2] during the daytime were found to be remarkably 318

greater under the 30%<RH<60% condition than under the RH≤30% condition, whereas the mean 319

values of [Dust] × [NO2] almost stayed the same under the two different RH conditions. 320

Considering the coincident trend of NOR and [HONO] × [DR] × [NO2] during the daytime 321

(10:00-17:00) under the 30%<RH<60% condition, the gas-phase reaction of NO2 with OH played 322

a key role in the diurnal formation of nitrate at moderate RH levels with the haze pollution 323

accumulating. It should be noted that the mean values of [HONO] × [DR] × [NO2] decreased 324

dramatically from 14:00 to 17:00 (Figure 3d), which was not responsible for the high mean values 325

of NOR at that time (Figure 3a). However, the slight increase in the mean values of [Dust] × [NO2] 326

after 14:00 was observed under the 30%<RH<60% condition (Figure 3c) and hence the 327

heterogeneous reaction of NO2 on the surface of mineral aerosols was suspected to contribute to 328

the diurnal formation of nitrate at that time under moderate RH condition. 329

3.3.2. Formation mechanism of sulfate 330

Atmospheric sulfate is principally from SO2 oxidation pathway, including gas-phase 331

reactions with OH radical or stabilized Criegee intermediates, heterogeneous-phase reactions on 332

the surface of particles and aqueous-phase reactions with dissolved O3, NO2, H2O2 and organic 333

peroxides, as well as autoxidation catalyzed by TMI (Cheng et al., 2016;Li et al., 334

2018;Ravishankara, 1997;Shao et al., 2019;Wang et al., 2016;Xue et al., 2016;Zhang et al., 2018). 335

Page 16: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

16

As shown in Figure 4, similar to the daily variations of NOR, the mean values of SOR were found 336

to elevated remarkably under the 30%<RH<60% condition compared to the RH30% condition, 337

especially during 14:00-22:00, which might be mainly ascribed to the enhanced gas-phase reaction 338

and the heterogeneous reaction of SO2 involving aerosol liquid water under the relatively high RH 339

condition. The extremely high mean values of SOR during the whole day under the RH≥60% 340

condition implied that aqueous oxidation of SO2 dominated the formation of sulfate during the 341

severe pollution episodes, which was in line with previous studies (Zhang et al., 2018;Cheng et al., 342

2016). A key factor that influenced the aqueous oxidation pathways for sulfate formation has been 343

considered to be the aerosol pH (Guo et al., 2017;Liu et al., 2017a), varying from 4.5 to 8.5 at 344

different atmospheric RH and sulfate levels during the sampling period (Figure 5a) on the basis of 345

the ISORROPIA-II model. Considering that the aqueous-phase chemistry of sulfate formation 346

usually occurs in severe haze events with relatively high atmospheric RH, the aerosol pH (4.5-5.3) 347

under the RH≥60% condition, which was lower than those (5.4-7.0) in the studies of Wang et al., 348

(2016) and Cheng et al., (2016) but was slightly higher than those (3.0-4.9) in the studies of Liu et 349

al., (2017a) and Guo et al., (2017), was adopted for evaluating sulfate production in this study. In 350

addition, in terms of oxidants, the obvious increase in the average concentration of NO2 (Figure 5b) 351

and the evident decrease in the average concentration of O3 (Figure 5d) were observed with the 352

deterioration of PM2.5 pollution. Furthermore, the average concentration of H2O2 was also found 353

to be extremely high (0.25 ppb) under the HP condition (Figure 5c) and was above 1 order of 354

magnitude higher than that (0.01 ppb) assumed by Cheng et al., (2016), which probably resulted in 355

the underestimation of the contribution of H2O2 to sulfate formation in the study of Cheng et al., 356

(2016). 357

Page 17: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

17

To further explore the contribution of H2O2 to sulfate production rate under the HP condition, 358

the parameters measured in this study (Table 2) and the same approach that was adopted by Cheng 359

et al., (2016) were used to calculate sulfate production. As shown in Figure 6, the relationships 360

between different aqueous oxidation pathways and aerosol pH in this study were found to be very 361

similar with those of Cheng et al., (2016). However, the contribution of H2O2 to sulfate production 362

rate was about a factor of 17 faster in this study (about 1.16 g m-3 h-1) than in the study (about 363

6.95×10-2 g m-3 h-1) of Cheng et al., (2016), implying that the contribution of H2O2 to sulfate 364

formation was largely neglected. Furthermore, considering the aerosol pH calculated under the HP 365

condition during the sampling period, the oxidation pathway of NO2 might play an insignificant 366

role in sulfate production rate (8.96×10-2-0.56 g m-3 h-1) and its importance proposed by the 367

previous studies (1.74-10.85 g m-3 h-1) was not necessarily expected. 368

4. Conclusion 369

Based on the comprehensive analysis of the pollution levels, the variation characteristics and 370

the formation mechanisms of the key species in PM2.5 and the typical gaseous pollutants during 371

the winter haze pollution periods in Beijing, three serious haze pollution cases were obtained 372

during the sampling period and the SIAs formations especially nitrate and sulfate were found to 373

make an evident contribution to atmospheric PM2.5 under the relatively high RH condition. The 374

significant correlation between [NO2]2 × [O3] and NOR at night under the RH≥60% condition 375

indicated that the heterogeneous hydrolysis of N2O5 on wet aerosols was responsible for the 376

nocturnal formation of nitrate under extremely high RH conditions. The more coincident trend of 377

NOR and [HONO] × [DR] × [NO2] than [Dust] × [NO2] during the daytime under the 378

30%<RH<60% condition suggested that the gas-phase reaction of NO2 with OH played a key role 379

Page 18: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

18

in the diurnal formation of nitrate under moderate RH conditions. The extremely high mean values 380

of SOR during the whole day under the RH≥60% condition could be explained by the dominant 381

contribution of aqueous-phase reaction of SO2 to atmospheric sulfate formation during the severe 382

pollution episodes. According to the parameters measured in this study and the same approach that 383

was adopted by Cheng et al., (2016), the oxidation pathway of H2O2 rather than NO2 was found to 384

contribute greatly to atmospheric sulfate formation. 385

Our results revealed that the heavy pollution events in winter usually occurred with high 386

concentration levels of pollutants and oxidants as well as high liquid water contents of moderately 387

acidic aerosols in the NCP. Thus, emission controls of NOx, SO2 and VOCs especially under the 388

extremely high RH conditions are expected to reduce largely the pollution levels of nitrate and 389

sulfate in northern China and even in other pollution regions of China. 390

391

Data availability. Data are available from the corresponding author upon request 392

([email protected]) 393

394

Author contributions. YJM designed the experiments. PFL carried out the experiments and 395

prepared the manuscript. CY and CYX carried out the experiments. CLZ was involved in part of 396

the work. XS provided the meteorological data and trace gases in Beijing. 397

398

Competing interests. The authors declare that they have no conflict of interest. 399

400

Acknowledgement. This work was supported by the National research program for Key issues in 401

Page 19: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

19

air pollution control (No. DQGG0103, DQGG0209, DQGG0206), the National Natural Science 402

Foundation of China (No. 91544211, 4127805, 41575121, 21707151), the National Key Research 403

and Development Program of China (No. 2016YFC0202200, 2017YFC0209703, 404

2017YFF0108301) and Key Laboratory of Atmospheric Chemistry, China Meteorological 405

Administration (No. 2018B03). 406

407

408

Figure 1. Time series of the species in PM2.5 and typical gaseous pollutants (NO2, SO2, O3, 409

HONO and H2O2) as well as atmospheric RH during the sampling period. 410

411

Page 20: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

20

412

Figure 2. The relations between NOR as well as SOR and RH during the sampling period. 413

414

415

Figure 3. Daily variation of NOR (a), the correlation between NOR and [NO2]2 × [O3] at the 416

nighttime (18:00-7:00) (b), daily variations of [Dust] × [NO2] and [HONO] × [DR] × [NO2] (c, d) 417

under different atmospheric RH conditions during the sampling period. 418

419

Page 21: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

21

420

Figure 4. Daily variation of SOR under different atmospheric RH conditions during the sampling 421

period. 422

423

424

Figure 5. The correlations among aerosol pH, atmospheric RH and atmospheric SO42- (a), the 425

average concentrations of NO2, H2O2 and O3 (b, c, d) under different pollution conditions (clean 426

(C), PM2.5<35 g m-3; slightly polluted (SP), 35 g m-3<PM2.5<75 g m-3; polluted (P), 75 g 427

Page 22: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

22

m-3<PM2.5<150 g m-3; heavy polluted (HP), PM2.5>150 g m-3) during the sampling period. 428

429

430

Figure 6. The comparison of aqueous-phase sulfate production by SO2 oxidation under different 431

aerosol pH conditions between in the study of Cheng et al., (2016) and in this study. 432

433

Table 1. The average concentrations of the species in PM2.5 (g m-3) and typical gaseous 434

pollutants (ppb) during C&SP episodes (PM2.5<75 g m-3), during P&HP episodes (PM2.5≥75 g 435

m-3) and during the whole sampling period. 436

species during C&SP episodes

(n=210)

during P&HP episodes

(n=108)

total

(n=318)

PM2.5 30.00 ± 17.79 113.35 ± 28.10 58.31 ± 45.15

Na+ 2.88 ± 1.11 3.68 ± 1.19 3.15 ± 1.21

Mg2+ 0.05 ± 0.03 0.08 ± 0.06 0.06 ± 0.04

Ca2+ 0.52 ± 0.33 0.67 ± 0.48 0.58 ± 0.40

K+ 0.81 ± 0.42 1.84 ± 0.73 1.16 ± 0.73

NH4+ 1.90 ± 1.90 11.52 ± 4.93 5.17 ± 5.62

SO42- 3.64 ± 1.87 14.96 ± 7.80 7.47 ± 7.18

NO3- 3.44 ± 3.57 17.15 ± 7.36 8.10 ± 8.32

Cl- 1.89 ± 1.20 7.35 ± 2.97 3.73 ± 3.26

NO2- 0.06 ± 0.08 0.06 ± 0.05 0.06 ± 0.07

OC 12.10 ± 9.25 43.34 ± 13.88 22.73 ± 18.48

EC 3.98 ± 3.42 12.69 ± 6.43 7.58 ± 6.51

NOx 39.38 ± 35.25 107.71 ± 58.44 62.59 ± 54.98

NO2 21.46 ± 13.04 42.81 ± 10.96 28.71 ± 15.98

SO2 6.99 ± 3.64 15.70 ± 6.55 9.95 ± 6.35

Page 23: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

23

O3 8.01 ± 6.35 2.13 ± 0.56 6.01 ± 5.87

HONO 0.60 ± 0.43 1.90 ± 0.97 1.01 ± 0.87

H2O2 0.17 ± 0.11 0.29 ± 0.14 0.20 ± 0.13

437

Table 2. The comparisons for parameters of sulfate production rate calculations between in the 438

study of Cheng et al., (2016) and in this work during the most polluted haze periods 439

Parameters This study Cheng et al., (2016)

NO2 57 ppb 66 ppb

H2O2 0.25 ppb 0.01 ppb

O3 2 ppb 1 ppb

SO2 35 ppb 40 ppb

Fe(III)a 18 ng m-3 18 ng m-3

Mn(II)a 42 ng m-3 42 ng m-3

ALWC 146 μg m-3 300 μg m-3

Aerosol droplet radius (R)a 0.15 μm 0.15 μm

Temperature 270 K 271 K

pH 4.5-5.3 5.4-6.2

a: both the concentrations of Fe(III) and Mn(II) and aerosol droplet radius were not measured in 440

this study and were derived from Cheng et al., (2016). 441

442

References 443

Bei, N., Wu, J., Elser, M., Feng, T., Cao, J., El-Haddad, I., Li, X., Huang, R., Li, Z., Long, X., Xing, L., 444

Zhao, S., Tie, X., Prévôt, A. S. H., and Li, G.: Impacts of meteorological uncertainties on the haze 445

formation in Beijing–Tianjin–Hebei (BTH) during wintertime: a case study, Atmospheric Chemistry 446

and Physics, 17, 14579-14591, 10.5194/acp-17-14579-2017, 2017. 447

Bougiatioti, A., Nikolaou, P., Stavroulas, I., Kouvarakis, G., Weber, R., Nenes, A., Kanakidou, M., and 448

Mihalopoulos, N.: Particle water and pH in the eastern Mediterranean: source variability and 449

implications for nutrient availability, Atmospheric Chemistry and Physics, 16, 4579-4591, 450

10.5194/acp-16-4579-2016, 2016. 451

Chan, C. K., and Yao, X.: Air pollution in mega cities in China, Atmospheric Environment, 42, 1-42, 452

10.1016/j.atmosenv.2007.09.003, 2008. 453

Chen, L. H., Sun, Y. Y., Wu, X. C., Zhang, Y. X., Zheng, C. H., Gao, X., and Cen, K.: Unit-based 454

emission inventory and uncertainty assessment of coal-fired power plants, Atmospheric Environment, 455

99, 527-535, 10.1016/j.atmosenv.2014.10.023, 2014. 456

Cheng, Y., Zheng, G., Wei, C., Mu, Q., Zheng, B., Wang, Z., Gao, M., Zhang, Q., He, K., Carmichael, 457

G., Pöschl, U., and Su, H.: Reactive nitrogen chemistry in aerosol water as a source of sulfate during 458

haze events in China, Science Advances, 2, 1-11, 10.1126/sciadv.1601530, 2016. 459

Clifton, C. L., Altstein, N., and Huie, R. E.: Rate-constant for the reaction of NO2 with sulfur(IV) over 460

the pH range 5.3-13, Environ. Sci. Technol., 22, 586-589, 10.1021/es00170a018, 1988. 461

Dai, Q., Bi, X., Song, W., Li, T., Liu, B., Ding, J., Xu, J., Song, C., Yang, N., Schulze, B. C., Zhang, Y., 462

Feng, Y., and Hopke, P. K.: Residential coal combustion as a source of primary sulfate in Xi'an, China, 463

Atmospheric Environment, 196, 66-76, 10.1016/j.atmosenv.2018.10.002, 2019. 464

Ding, J., Zhao, P., Su, J., Dong, Q., Du, X., and Zhang, Y.: Aerosol pH and its driving factors in Beijing, 465

Page 24: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

24

Atmospheric Chemistry and Physics, 19, 7939-7954, 10.5194/acp-19-7939-2019, 2019. 466

Du, Q., Zhang, C., Mu, Y., Cheng, Y., Zhang, Y., Liu, C., Song, M., Tian, D., Liu, P., Liu, J., Xue, C., 467

and Ye, C.: An important missing source of atmospheric carbonyl sulfide: Domestic coal combustion, 468

Geophysical Research Letters, 43, 8720-8727, 10.1002/2016gl070075, 2016. 469

Fountoukis, C., and Nenes, A.: ISORROPIA II: a computationally efficient thermodynamic equilibrium 470

model for K+-Ca2+-Mg2+-NH4+-Na+-SO4

2--NO3--Cl--H2O aerosols, Atmospheric Chemistry and Physics, 471

7, 4639-4659, 2007. 472

Ge, X., He, Y., Sun, Y., Xu, J., Wang, J., Shen, Y., and Chen, M.: Characteristics and Formation 473

Mechanisms of Fine Particulate Nitrate in Typical Urban Areas in China, Atmosphere, 8, 62, 474

10.3390/atmos8030062, 2017. 475

Graedel, T. E., and Weschler, C. J.: Chemistry within aqueous atmospheric aerosols and raindrops, 476

Reviews of Geophysics, 19, 505-539, 10.1029/RG019i004p00505, 1981. 477

Guo, H., Xu, L., Bougiatioti, A., Cerully, K. M., Capps, S. L., Hite, J. R., Carlton, A. G., Lee, S. H., 478

Bergin, M. H., Ng, N. L., Nenes, A., and Weber, R. J.: Fine-particle water and pH in the southeastern 479

United States, Atmospheric Chemistry and Physics, 15, 5211-5228, 10.5194/acp-15-5211-2015, 2015. 480

Guo, H., Weber, R. J., and Nenes, A.: High levels of ammonia do not raise fine particle pH sufficiently 481

to yield nitrogen oxide-dominated sulfate production, Scientific reports, 7, 12109, 482

10.1038/s41598-017-11704-0, 2017. 483

Guo, S., Hu, M., Zamora, M. L., Peng, J., Shang, D., Zheng, J., Du, Z., Wu, Z., Shao, M., Zeng, L., 484

Molina, M. J., and Zhang, R.: Elucidating severe urban haze formation in China, Proceedings of the 485

National Academy of Sciences of the United States of America, 111, 17373-17378, 486

10.1073/pnas.1419604111, 2014. 487

He, H., Wang, Y., Ma, Q., Ma, J., Chu, B., Ji, D., Tang, G., Liu, C., Zhang, H., and Hao, J.: Mineral 488

dust and NOx promote the conversion of SO2 to sulfate in heavy pollution days, Scientific reports, 4, 489

1-5, 10.1038/srep04172, 2014. 490

He, P., Xie, Z., Chi, X., Yu, X., Fan, S., Kang, H., Liu, C., and Zhan, H.: Atmospheric Δ17O(NO3−) 491

reveals nocturnal chemistry dominates nitrate production in Beijing haze, Atmospheric Chemistry and 492

Physics, 18, 14465-14476, 10.5194/acp-18-14465-2018, 2018. 493

Hennigan, C. J., Izumi, J., Sullivan, A. P., Weber, R. J., and Nenes, A.: A critical evaluation of proxy 494

methods used to estimate the acidity of atmospheric particles, Atmospheric Chemistry and Physics, 15, 495

2775-2790, 10.5194/acp-15-2775-2015, 2015. 496

Huang, R. J., Zhang, Y., Bozzetti, C., Ho, K. F., Cao, J. J., Han, Y., Daellenbach, K. R., Slowik, J. G., 497

Platt, S. M., Canonaco, F., Zotter, P., Wolf, R., Pieber, S. M., Bruns, E. A., Crippa, M., Ciarelli, G., 498

Piazzalunga, A., Schwikowski, M., Abbaszade, G., Schnelle-Kreis, J., Zimmermann, R., An, Z., Szidat, 499

S., Baltensperger, U., El Haddad, I., and Prevot, A. S.: High secondary aerosol contribution to 500

particulate pollution during haze events in China, Nature, 514, 218-222, 10.1038/nature13774, 2014. 501

Ibusuki, T., and Takeuchi, K.: Sulfur-dioxide oxidation by oxygen catalyzed by mixtures of 502

manganese(II) and iron(III) in aqueous-solutions at environmental reaction conditions, Atmospheric 503

Environment, 21, 1555-1560, 10.1016/0004-6981(87)90317-9, 1987. 504

Li, G., Bei, N., Cao, J., Huang, R., Wu, J., Feng, T., Wang, Y., Liu, S., Zhang, Q., Tie, X., and Molina, 505

L. T.: A possible pathway for rapid growth of sulfate during haze days in China, Atmospheric 506

Chemistry and Physics, 17, 3301-3316, 10.5194/acp-17-3301-2017, 2017. 507

Li, J., Liao, H., Hu, J., and Li, N.: Severe particulate pollution days in China during 2013-2018 and the 508

associated typical weather patterns in Beijing-Tianjin-Hebei and the Yangtze River Delta regions, 509

Page 25: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

25

Environ Pollut, 248, 74-81, 10.1016/j.envpol.2019.01.124, 2019. 510

Li, L., Hoffmann, M. R., and Colussi, A. J.: Role of nitrogen dioxide in the production of sulfate during 511

Chinese haze-aerosol episodes, Environ Sci Technol, 52, 2686-2693, 10.1021/acs.est.7b05222, 2018. 512

Li, Q., Li, X., Jiang, J., Duan, L., Ge, S., Zhang, Q., Deng, J., Wang, S., and Hao, J.: Semi-coke 513

briquettes: towards reducing emissions of primary PM2.5, particulate carbon, and carbon monoxide 514

from household coal combustion in China, Scientific reports, 6, 1-10, 10.1038/srep19306, 2016. 515

Liu, M., Song, Y., Zhou, T., Xu, Z., Yan, C., Zheng, M., Wu, Z., Hu, M., Wu, Y., and Zhu, T.: Fine 516

particle pH during severe haze episodes in northern China, Geophysical Research Letters, 44, 1-9, 517

10.1002/2017GL073210, 2017a. 518

Liu, P., Zhang, C., Mu, Y., Liu, C., Xue, C., Ye, C., Liu, J., Zhang, Y., and Zhang, H.: The possible 519

contribution of the periodic emissions from farmers' activities in the North China Plain to atmospheric 520

water-soluble ions in Beijing, Atmospheric Chemistry and Physics, 16, 10097-10109, 521

10.5194/acp-16-10097-2016, 2016. 522

Liu, P., Zhang, C., Xue, C., Mu, Y., Liu, J., Zhang, Y., Tian, D., Ye, C., Zhang, H., and Guan, J.: The 523

contribution of residential coal combustion to atmospheric PM2.5 in northern China during winter, 524

Atmospheric Chemistry and Physics, 17, 11503-11520, 10.5194/acp-17-11503-2017, 2017b. 525

Ma, Q., Wang, T., Liu, C., He, H., Wang, Z., Wang, W., and Liang, Y.: SO2 Initiates the efficient 526

conversion of NO2 to HONO on MgO Surface, Environ Sci Technol, 51, 3767-3775, 527

10.1021/acs.est.6b05724, 2017. 528

Meng, Z. Y., Lin, W. L., Jiang, X. M., Yan, P., Wang, Y., Zhang, Y. M., Jia, X. F., and Yu, X. L.: 529

Characteristics of atmospheric ammonia over Beijing, China, Atmospheric Chemistry and Physics, 11, 530

6139-6151, 10.5194/acp-11-6139-2011, 2011. 531

Murphy, J. G., Gregoire, P. K., Tevlin, A. G., Wentworth, G. R., Ellis, R. A., Markovic, M. Z., and 532

VandenBoer, T. C.: Observational constraints on particle acidity using measurements and modelling of 533

particles and gases, Faraday Discussions, 200, 379-395, 10.1039/c7fd00086c, 2017. 534

Nie, W., Ding, A., Wang, T., Kerminen, V. M., George, C., Xue, L., Wang, W., Zhang, Q., Petaja, T., Qi, 535

X., Gao, X., Wang, X., Yang, X., Fu, C., and Kulmala, M.: Polluted dust promotes new particle 536

formation and growth, Scientific reports, 4, 1-6, 10.1038/srep06634, 2014. 537

Pathak, R. K., Louie, P. K. K., and Chan, C. K.: Characteristics of aerosol acidity in Hong kong, 538

Atmospheric Environment, 38, 2965-2974, 10.1016/j.atmosenv.2004.02.044, 2004. 539

Ponczek, M., Hayeck, N., Emmelin, C., and George, C.: Heterogeneous photochemistry of dicarboxylic 540

acids on mineral dust, Atmospheric Environment, 212, 262-271, 10.1016/j.atmosenv.2019.05.032, 541

2019. 542

Quan, J., Tie, X., Zhang, Q., Liu, Q., Li, X., Gao, Y., and Zhao, D.: Characteristics of heavy aerosol 543

pollution during the 2012–2013 winter in Beijing, China, Atmospheric Environment, 88, 83-89, 544

10.1016/j.atmosenv.2014.01.058, 2014. 545

Ravishankara, A.: Heterogeneous and multiphase chemistry in the troposphere, Science, 276, 546

1058-1065, 1997. 547

Seinfeld, J. H., and Pandis, S. N.: Atmospheric Chemistry and Physics, from Air Pollution to Climate 548

Change, Wiley, 429-443 pp., 2006. 549

Shao, J., Chen, Q., Wang, Y., Lu, X., He, P., Sun, Y., Shah, V., Martin, R. V., Philip, S., Song, S., Zhao, 550

Y., Xie, Z., Zhang, L., and Alexander, B.: Heterogeneous sulfate aerosol formation mechanisms during 551

wintertime Chinese haze events: air quality model assessment using observations of sulfate oxygen 552

isotopes in Beijing, Atmospheric Chemistry and Physics, 19, 6107-6123, 10.5194/acp-19-6107-2019, 553

Page 26: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

26

2019. 554

Shi, G., Xu, J., Peng, X., Xiao, Z., Chen, K., Tian, Y., Guan, X., Feng, Y., Yu, H., Nenes, A., and 555

Russell, A. G.: pH of aerosols in a polluted atmosphere: source contributions to highly acidic aerosol, 556

Environ Sci Technol, 51, 4289-4296, 10.1021/acs.est.6b05736, 2017. 557

Sun, Y., Wang, Z., Fu, P., Jiang, Q., Yang, T., Li, J., and Ge, X.: The impact of relative humidity on 558

aerosol composition and evolution processes during wintertime in Beijing, China, Atmospheric 559

Environment, 77, 927-934, 10.1016/j.atmosenv.2013.06.019, 2013. 560

Tang, M., Huang, X., Lu, K., Ge, M., Li, Y., Cheng, P., Zhu, T., Ding, A., Zhang, Y., Gligorovski, S., 561

Song, W., Ding, X., Bi, X., and Wang, X.: Heterogeneous reactions of mineral dust aerosol: 562

implications for tropospheric oxidation capacity, Atmospheric Chemistry and Physics, 17, 11727-11777, 563

10.5194/acp-17-11727-2017, 2017. 564

Tham, Y. J., Wang, Z., Li, Q., Wang, W., Wang, X., Lu, K., Ma, N., Yan, C., Kecorius, S., Wiedensohler, 565

A., Zhang, Y., and Wang, T.: Heterogeneous N2O5 uptake coefficient and production yield of ClNO2 in 566

polluted northern China: roles of aerosol water content and chemical composition, Atmospheric 567

Chemistry and Physics, 18, 13155-13171, 10.5194/acp-18-13155-2018, 2018. 568

Tong, S. R., Hou, S. Q., Zhang, Y., Chu, B. W., Liu, Y. C., He, H., Zhao, P. S., and Ge, M. F.: Exploring 569

the nitrous acid (HONO) formation mechanism in winter Beijing: direct emissions and heterogeneous 570

production in urban and suburban areas, Faraday Discussions, 189, 213-230, 10.1039/c5fd00163c, 571

2016. 572

Wang, G., Zhang, R., Gomez, M. E., Yang, L., Levy Zamora, M., Hu, M., Lin, Y., Peng, J., Guo, S., 573

Meng, J., Li, J., Cheng, C., Hu, T., Ren, Y., Wang, Y., Gao, J., Cao, J., An, Z., Zhou, W., Li, G., Wang, 574

J., Tian, P., Marrero-Ortiz, W., Secrest, J., Du, Z., Zheng, J., Shang, D., Zeng, L., Shao, M., Wang, W., 575

Huang, Y., Wang, Y., Zhu, Y., Li, Y., Hu, J., Pan, B., Cai, L., Cheng, Y., Ji, Y., Zhang, F., Rosenfeld, D., 576

Liss, P. S., Duce, R. A., Kolb, C. E., and Molina, M. J.: Persistent sulfate formation from London Fog 577

to Chinese haze, Proceedings of the National Academy of Sciences of the United States of America, 578

113, 13630-13635, 2016. 579

Wang, G., Zhang, F., Peng, J., Duan, L., Ji, Y., Marrero-Ortiz, W., Wang, J., Li, J., Wu, C., Cao, C., 580

Wang, Y., Zheng, J., Secrest, J., Li, Y., Wang, Y., Li, H., Li, N., and Zhang, R.: Particle acidity and 581

sulfate production during severe haze events in China cannot be reliably inferred by assuming a 582

mixture of inorganic salts, Atmospheric Chemistry and Physics, 18, 10123-10132, 583

10.5194/acp-18-10123-2018, 2018a. 584

Wang, H., Lu, K., Chen, X., Zhu, Q., Wu, Z., Wu, Y., and Sun, K.: Fast particulate nitrate formation via 585

N2O5 uptake aloft in winter in Beijing, Atmospheric Chemistry and Physics, 18, 10483-10495, 586

10.5194/acp-18-10483-2018, 2018b. 587

Wang, H., Lu, K., Guo, S., Wu, Z., Shang, D., Tan, Z., Wang, Y., Le Breton, M., Lou, S., Tang, M., Wu, 588

Y., Zhu, W., Zheng, J., Zeng, L., Hallquist, M., Hu, M., and Zhang, Y.: Efficient N2O5 uptake and NO3 589

oxidation in the outflow of urban Beijing, Atmospheric Chemistry and Physics, 18, 9705-9721, 590

10.5194/acp-18-9705-2018, 2018c. 591

Wang, J., Zhang, X., Guo, J., Wang, Z., and Zhang, M.: Observation of nitrous acid (HONO) in Beijing, 592

China: Seasonal variation, nocturnal formation and daytime budget, The Science of the total 593

environment, 587-588, 350-359, 10.1016/j.scitotenv.2017.02.159, 2017. 594

Wang, Y., Yao, L., Wang, L., Liu, Z., Ji, D., Tang, G., Zhang, J., Sun, Y., Hu, B., and Xin, J.: 595

Mechanism for the formation of the January 2013 heavy haze pollution episode over central and 596

eastern China, Science China Earth Sciences, 57, 14-25, 10.1007/s11430-013-4773-4, 2013. 597

Page 27: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

27

Weber, R. J., Guo, H., Russell, A. G., and Nenes, A.: High aerosol acidity despite declining 598

atmospheric sulfate concentrations over the past 15 years, Nature Geoscience, 9, 282-285, 599

10.1038/ngeo2665, 2016. 600

Wu, J., Bei, N., Hu, B., Liu, S., Zhou, M., Wang, Q., Li, X., Liu, L., Feng, T., Liu, Z., Wang, Y., Cao, J., 601

Tie, X., Wang, J., Molina, L. T., and Li, G.: Is water vapor a key player of the wintertime haze in North 602

China Plain?, Atmospheric Chemistry and Physics, 19, 8721-8739, 10.5194/acp-19-8721-2019, 2019. 603

Xu, L., Duan, F., He, K., Ma, Y., Zhu, L., Zheng, Y., Huang, T., Kimoto, T., Ma, T., Li, H., Ye, S., Yang, 604

S., Sun, Z., and Xu, B.: Characteristics of the secondary water-soluble ions in a typical autumn haze in 605

Beijing, Environ Pollut, 227, 296-305, 10.1016/j.envpol.2017.04.076, 2017. 606

Xu, W. Y., Zhao, C. S., Ran, L., Deng, Z. Z., Liu, P. F., Ma, N., Lin, W. L., Xu, X. B., Yan, P., He, X., 607

Yu, J., Liang, W. D., and Chen, L. L.: Characteristics of pollutants and their correlation to 608

meteorological conditions at a suburban site in the North China Plain, Atmospheric Chemistry and 609

Physics, 11, 4353-4369, 10.5194/acp-11-4353-2011, 2011. 610

Xue, C., Ye, C., Ma, Z., Liu, P., Zhang, Y., Zhang, C., Tang, K., Zhang, W., Zhao, X., Wang, Y., Song, 611

M., Liu, J., Duan, J., Qin, M., Tong, S., Ge, M., and Mu, Y.: Development of stripping coil-ion 612

chromatograph method and intercomparison with CEAS and LOPAP to measure atmospheric HONO, 613

The Science of the total environment, 646, 187-195, 10.1016/j.scitotenv.2018.07.244, 2019a. 614

Xue, C., Ye, C., Zhang, Y., Ma, Z., Liu, P., Zhang, C., Zhao, X., Liu, J., and Mu, Y.: Development and 615

application of a twin open-top chambers method to measure soil HONO emission in the North China 616

Plain, Sci. Total Environ., 659, 621-631, 10.1016/j.scitotenv.2018.12.245, 2019b. 617

Xue, J., Yuan, Z., Griffith, S. M., Yu, X., Lau, A. K., and Yu, J. Z.: Sulfate Formation Enhanced by a 618

Cocktail of High NOx, SO2, Particulate Matter, and Droplet pH during Haze-Fog Events in Megacities 619

in China: An Observation-Based Modeling Investigation, Environ Sci Technol, 50, 7325-7334, 620

10.1021/acs.est.6b00768, 2016. 621

Yang, T., Sun, Y., Zhang, W., Wang, Z., Liu, X., Fu, P., and Wang, X.: Evolutionary processes and 622

sources of high-nitrate haze episodes over Beijing, Spring, J Environ Sci (China), 54, 142-151, 623

10.1016/j.jes.2016.04.024, 2017. 624

Yang, Y. R., Liu, X. G., Qu, Y., An, J. L., Jiang, R., Zhang, Y. H., Sun, Y. L., Wu, Z. J., Zhang, F., Xu, 625

W. Q., and Ma, Q. X.: Characteristics and formation mechanism of continuous hazes in China: a case 626

study during the autumn of 2014 in the North China Plain, Atmospheric Chemistry and Physics, 15, 627

8165-8178, 10.5194/acp-15-8165-2015, 2015. 628

Ye, C., Liu, P., Ma, Z., Xue, C., Zhang, C., Zhang, Y., Liu, J., Liu, C., Sun, X., and Mu, Y.: High H2O2 629

Concentrations Observed during Haze Periods during the Winter in Beijing: Importance of H2O2 630

Oxidation in Sulfate Formation, Environmental Science & Technology Letters, 5, 757-763, 631

10.1021/acs.estlett.8b00579, 2018. 632

Zhang, H., Chen, S., Zhong, J., Zhang, S., Zhang, Y., Zhang, X., Li, Z., and Zeng, X. C.: Formation of 633

aqueous-phase sulfate during the haze period in China: Kinetics and atmospheric implications, 634

Atmospheric Environment, 177, 93-99, 10.1016/j.atmosenv.2018.01.017, 2018. 635

Zhang, Q., He, K. B., and Huo, H.: Cleaning China's air, Nature, 484, 161-162, 2012. 636

Zhang, R., Wang, G., Guo, S., Zamora, M. L., Ying, Q., Lin, Y., Wang, W., Hu, M., and Wang, Y.: 637

Formation of urban fine particulate matter, Chem Rev, 115, 3803-3855, 10.1021/acs.chemrev.5b00067, 638

2015. 639

Zhao, M., Wang, S., Tan, J., Hua, Y., Wu, D., and Hao, J.: Variation of Urban Atmospheric Ammonia 640

Pollution and its Relation with PM2.5 Chemical Property in Winter of Beijing, China, Aerosol and Air 641

Page 28: Formation mechanisms of atmospheric nitrate and sulfate ... · 73 proposed that atmospheric SO2 oxidation by NO2 dissolved in aqueous aerosol phases under the 74 extremely high atmospheric

28

Quality Research, 16, 1390-1402, 10.4209/aaqr.2015.12.0699, 2016. 642

Zheng, B., Zhang, Q., Zhang, Y., He, K. B., Wang, K., Zheng, G. J., Duan, F. K., Ma, Y. L., and Kimoto, 643

T.: Heterogeneous chemistry: a mechanism missing in current models to explain secondary inorganic 644

aerosol formation during the January 2013 haze episode in North China, Atmospheric Chemistry and 645

Physics, 15, 2031-2049, 10.5194/acp-15-2031-2015, 2015a. 646

Zheng, G. J., Duan, F. K., Su, H., Ma, Y. L., Cheng, Y., Zheng, B., Zhang, Q., Huang, T., Kimoto, T., 647

Chang, D., Pöschl, U., Cheng, Y. F., and He, K. B.: Exploring the severe winter haze in Beijing: the 648

impact of synoptic weather, regional transport and heterogeneous reactions, Atmospheric Chemistry 649

and Physics, 15, 2969-2983, 10.5194/acp-15-2969-2015, 2015b. 650

Zhong, J., Zhang, X., Wang, Y., Wang, J., Shen, X., Zhang, H., Wang, T., Xie, Z., Liu, C., Zhang, H., 651

Zhao, T., Sun, J., Fan, S., Gao, Z., Li, Y., and Wang, L.: The two-way feedback mechanism between 652

unfavorable meteorological conditions and cumulative aerosol pollution in various haze regions of 653

China, Atmospheric Chemistry and Physics, 19, 3287-3306, 10.5194/acp-19-3287-2019, 2019. 654

655